Ethan
commited on
Commit
·
22be7da
1
Parent(s):
bd80e50
two
Browse files- .gitattributes +7 -0
- README.md +3 -3
- adapter_config.json +3 -31
- training_log.json +3 -16
- training_parameters.json +3 -37
- training_prompt.json +3 -3
.gitattributes
CHANGED
@@ -33,3 +33,10 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
training_log.json filter=lfs diff=lfs merge=lfs -text
|
37 |
+
training_parameters.json filter=lfs diff=lfs merge=lfs -text
|
38 |
+
training_prompt.json filter=lfs diff=lfs merge=lfs -text
|
39 |
+
README.md filter=lfs diff=lfs merge=lfs -text
|
40 |
+
adapter_config.json filter=lfs diff=lfs merge=lfs -text
|
41 |
+
adapter_model.safetensors filter=lfs diff=lfs merge=lfs -text
|
42 |
+
runs filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:98b45ea81164d1e1a1dd82255207053b15cd6c69d922a1c5cf3387ce604d4b74
|
3 |
+
size 28
|
adapter_config.json
CHANGED
@@ -1,31 +1,3 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
"base_model_name_or_path": "models/TheBloke_speechless-mistral-dolphin-orca-platypus-samantha-7B-GPTQ",
|
5 |
-
"bias": "none",
|
6 |
-
"fan_in_fan_out": false,
|
7 |
-
"inference_mode": true,
|
8 |
-
"init_lora_weights": true,
|
9 |
-
"layers_pattern": null,
|
10 |
-
"layers_to_transform": null,
|
11 |
-
"loftq_config": {},
|
12 |
-
"lora_alpha": 64,
|
13 |
-
"lora_dropout": 0.05,
|
14 |
-
"megatron_config": null,
|
15 |
-
"megatron_core": "megatron.core",
|
16 |
-
"modules_to_save": null,
|
17 |
-
"peft_type": "LORA",
|
18 |
-
"r": 32,
|
19 |
-
"rank_pattern": {},
|
20 |
-
"revision": null,
|
21 |
-
"target_modules": [
|
22 |
-
"down_proj",
|
23 |
-
"o_proj",
|
24 |
-
"k_proj",
|
25 |
-
"q_proj",
|
26 |
-
"up_proj",
|
27 |
-
"v_proj",
|
28 |
-
"gate_proj"
|
29 |
-
],
|
30 |
-
"task_type": "CAUSAL_LM"
|
31 |
-
}
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2365a8875421a2e108fc0f2c201e7772bc8c27a4ec3d0a1aa01325e4984097c2
|
3 |
+
size 703
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
training_log.json
CHANGED
@@ -1,16 +1,3 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
"base_loaded_in_4bit": false,
|
5 |
-
"base_loaded_in_8bit": false,
|
6 |
-
"projections": "gate, down, up, q, v, k, o",
|
7 |
-
"train_runtime": 504.3907,
|
8 |
-
"train_samples_per_second": 19.826,
|
9 |
-
"train_steps_per_second": 0.099,
|
10 |
-
"total_flos": 2115741155328000.0,
|
11 |
-
"train_loss": 4.443607044219971,
|
12 |
-
"epoch": 32.0,
|
13 |
-
"current_steps": 1585,
|
14 |
-
"loss": 4.0742,
|
15 |
-
"learning_rate": 0.0
|
16 |
-
}
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ceb2ef2aadfbbf6eaf78281da4a97672219b5a9247f3445ce2ed3bc4da17864b
|
3 |
+
size 505
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
training_parameters.json
CHANGED
@@ -1,37 +1,3 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
"q_proj_en": true,
|
5 |
-
"v_proj_en": true,
|
6 |
-
"k_proj_en": true,
|
7 |
-
"o_proj_en": true,
|
8 |
-
"gate_proj_en": true,
|
9 |
-
"down_proj_en": true,
|
10 |
-
"up_proj_en": true,
|
11 |
-
"save_steps": 0.0,
|
12 |
-
"micro_batch_size": 4,
|
13 |
-
"batch_size": 128,
|
14 |
-
"epochs": 50.0,
|
15 |
-
"learning_rate": "3e-4",
|
16 |
-
"lr_scheduler_type": "linear",
|
17 |
-
"lora_rank": 32,
|
18 |
-
"lora_alpha": 64,
|
19 |
-
"lora_dropout": 0.05,
|
20 |
-
"cutoff_len": 256,
|
21 |
-
"dataset": "None",
|
22 |
-
"eval_dataset": "None",
|
23 |
-
"format": "None",
|
24 |
-
"eval_steps": 100.0,
|
25 |
-
"raw_text_file": "50-elijah",
|
26 |
-
"overlap_len": 128,
|
27 |
-
"newline_favor_len": 128,
|
28 |
-
"higher_rank_limit": false,
|
29 |
-
"warmup_steps": 100.0,
|
30 |
-
"optimizer": "adamw_torch",
|
31 |
-
"hard_cut_string": "\\n\\n",
|
32 |
-
"train_only_after": "",
|
33 |
-
"stop_at_loss": 0,
|
34 |
-
"add_eos_token": false,
|
35 |
-
"min_chars": 0.0,
|
36 |
-
"report_to": "None"
|
37 |
-
}
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:082d7caa0b6e26a4cf6a13d9beca159e92a0d52abc34e07535c7d147964e63ba
|
3 |
+
size 847
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
training_prompt.json
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f9fe797a498b6def105ea1e70735d61009194922c5ab5e0a74a2e427797a5e75
|
3 |
+
size 33
|